Transitional Tholeiitic Basalts in the Tertiary Bana Volcano–Plutonic Complex, Cameroon Line

Home , Alkali basalt, Basanite, Benmoreite, Hawaiite

Journal of African Earth Sciences 45 (2006) 318–332 www.elsevier.com/locate/jafrearsci

Transitional tholeiitic basalts in the Tertiary Bana volcano–plutonic complex, Cameroon Line

Gilbert Kuepouo a,b,*, Jean Pierre Tchouankoue b, Takashi Nagao c, Hiroaki Sato a

a Graduate School of Science and Technology, Department of Earth and Planetary Sciences, Faculty of Science, Kobe University, Nada, Kobe 657-8501, Japan b Department of Earth Sciences, Internal Geodynamics Laboratory, University of Yaounde-I, PO Box 812, Yaounde, Cameroon c Center for Instrumental Analysis, Yamaguchi University, 1677-1 Yoshida, Yamaguchi 753-8512, Japan

Received 10 May 2005; received in revised form 7 March 2006; accepted 9 March 2006 Available online 18 May 2006

Abstract

The Bana transitional tholeiitic basalts occurring in a Tertiary volcano–plutonic complex of the Cameroon Line, Central Africa are plagioclase-bearing and olivine-free. K/Ar dating on separated plagioclases of the transitional tholeiitic basalts yields an Oligocene age of 30.1 ± 1.2 Ma. Their clinopyroxene compositions are marked by iron enrichment and calcium depletion in the Wo–En–Fs system. The whole-rock major element compositions are characterized by Mg# 36–48, normative quartz and hypersthene. The youngest alkali basalts from the same igneous complex have higher Mg# 56–66. These two groups of basalt have trace element characteristics of within- plate basalt with Zr/Nb ratios of 3.7–4.5 and 7.5–9.2 respectively, and diﬀerent LILE/HFSE and LREE/HREE ratios. The overall trace element characteristics suggest that the transitional tholeiitic basalts of the Bana complex were derived by high degrees of partial melting in the upper mantle at shallow depths whereas younger alkali basalts in the complex were probably produced by a small degree of melting of the same source at slightly greater depths. The transitional tholeiitic character of these basalts suggests a signiﬁcant lithospheric extension and mantle upwelling below the Cameroon Line in the Oligocene. Ó 2006 Elsevier Ltd. All rights reserved.

Keywords: Transitional tholeiitic basalts; WPB; Tertiary Bana volcano–plutonic complex; Cameroon Line

1. Introduction uted to the understanding of geochemical and isotopic features of alkali basalts (Fitton and Dunlop, 1985; Fitton, Oceanic and continental lavas constitute the extrusive 1987; Halliday et al., 1988, 1990; Lee et al., 1994; Marzoli section of the Cameroon Line (Fig. 1), a line of Eocene– et al., 2000; Rankenburg et al., 2005). These studies high- Oligocene anorogenic volcano–plutonic complexes, and lighted the chemical and isotopic similarities between bas- Oligocene to Present volcanic centers. alts from oceanic and continental sectors, their alkaline The Cameroon Line comprises about 60 anorogenic vol- nature and the dominance of the HIMU mantle source cano–plutonic complexes, and a large number of polyge- during their genesis (Halliday et al., 1990). netic and monogenetic volcanoes extending from Pagalu De´ruelle et al. (1991) showed that extensive petrological Island in the Atlantic Ocean (SW) to lake Chad (NE) on studies of the Cameroon Line volcanic and plutonic rocks the Africa continent. A number of studies of continental do not allow the conclusion that magmatism develops a and oceanic basalts of the Cameroon Line have contrib- transitional trend in that region as is the case in the East- African Rift. They discarded a transitional character of some basalts from Mt Oku, Manengouba and Principe * Corresponding author. Address: Department of Earth Sciences, Internal Geodynamics Laboratory, University of Yaounde-I, PO Box island (Fitton, 1987), and gabbros from the Mboutou 812, Yaounde, Cameroon. Tel.: +237 720 22 71. igneous complex (Parsons et al., 1986) on the basis of their E-mail address: [email protected] (G. Kuepouo). high niobium contents compared with typical transitional

Fig. 1. Geologic map of the Bana plutono–volcanic complex (Kuepouo et al., 2004). Inset shows the position of the Bana volcano–plutonic complex in the Cameroon line. basalts of the East-African Rift (Kampunzu and Mohr, basalts, and typified by their mildly hypersthene-normative 1991) and OIB. Further record of transitional basalts to mildly nepheline-normative character. within the Cameroon Line by some workers (Kampunzu The objectives of this paper are: (1) to prove the occur- and Lubala, 1991; Moundi et al., 1996; Moundi, 2004; rence of transitional tholeiitic basalts discovered within the Fosso et al., 2005) is hotly debated. Cameroon Line; (2) to discuss their clinopyroxene and For clarification, transitional basalts are basalts having whole-rock chemistries in comparison with those of typical compositions intermediate between tholeiitic and alkaline alkali basalts of the Cameroon Line occurring in the same 320 G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332 complex; and (3) to define and understand the genetic rela- Previous petrological studies of basalts along the Cam- tionship between these two groups of basalts in this setting. eroon Line showed that they were derived from a depleted asthenospheric source beneath the subcontinental litho- 2. Geological background sphere (Halliday et al., 1988; Sato et al., 1990), or from amphibole-bearing lithospheric mantle for the continental The Tertiary Bana volcano–plutonic complex (5°80S, basalts (Marzoli et al., 2000). Geophysical studies revealed 5°110N; area ca. 55 km2) is located southeast of the West that the Cameroon Line is underlain by a thin crust of Cameroon Highlands in the central part of the Cameroon ca. 30–34 km thickness (Fairhead and Okereke, 1987; Line (Fig. 1). A previous petrological study on the Bana Plomerova et al., 1993; Poudjom Djomani et al., 1995). complex (Nana, 1988) distinguished: (1) the plutonic Previous hypotheses on the origin of the Cameroon Line unit including a biotite ± amphibole granite, an arfvedso- are documented elsewhere (Moreau et al., 1987; De´ruelle nite ± aegirine granite, and a small lens of leucogabbro; et al., 1991; Burke, 2001). Recent studies by Poudjom (2) the volcanic unit made up of small lava flows, basaltic Djomani et al. (1997) suggest that the Cameroon Line fol- lapilli tuffs and rhyolitic cinder tuffs forming a ring around lows a major structural zone in the lithosphere. and on top of the arfvedsonite ± aegirine granite at the North. 3. Petrography Attempts to date the Bana complex by the K/Ar method resulted in values that ranged from 38 ± 1 and 42 ± 8 Ma 3.1. Plagioclase-basalts on benmoreite lavas (Cantagrel et al., 1978; Nana, 1988) and 30 Ma (no analytical precision reported) on ‘‘evolved The plagioclase-basalts from Bana are subaphyric to lava’’ (Lasserre, 1978). The arfvedsonite ± aegirine granite porphyritic with a primary mineral assemblage dominated yielded an Rb/Sr age of 51 ± 1 Ma and an initial 87Sr/86Sr by plagioclase, Fe–Ti oxides and clinopyroxene. In plagio- ratio of 0.7035 ± 0.0001 (Caen-Vachette et al., 1991). clase-phyric basalts, plagioclase phenocrysts >3 mm across A detailed geologic map (Fig. 1) of the Bana volcano– represent ca. 30–50 vol.%. Large plagioclase phenocrysts plutonic complex is available from Kuepouo (2004). The range from 5 to 15 mm in length. Individual phenocrysts Bana complex forms a prominent mountainous scarp cul- are euhedral to subhedral, although strongly corroded minating at 2097 m above sea level, rising 600 m above crystals were observed. Glomeroporphyritic associations the general level of the countryside to the south and show cruciform intergrowth of plagioclase. The intersertal 350 m to the northwest. This complex is bounded to the to intergranular groundmass is dominated by plagioclase south by an elevated (up to 1500 m) crystalline basement microlites. Phenocryst and microlite compositions vary encompassing Neoproterozoic granite and gneisses, cross from bytownite to andesine. Interstitial minerals are cut by mafic and felsic dykes. Aerial photographs and field Fe–Ti oxides, apatite needles and occasionally albite and observations reveal a sharp discordant contact between the accessory titanite. Plagioclase phenocrysts are partly saus- complex and the basement to the south. suritized in altered samples and the groundmass partly The complex includes two crescent-like plutonic units replaced by micas, epidote and carbonate. Chlorite replaces forming the ‘‘Southern Intrusions’’ and the ‘‘Northern glass. Intrusions’’ (Fig. 1). From west to east, the Southern Intru- Clinopyroxene constitutes 7.6 vol.% mostly as microlites sions consist of biotite-amphibole granite and biotite (<0.3 mm) and occasional microphenocrysts (up to 1 mm miarolitic granite, whereas the Northern Intrusions are across). Microphenocrysts form isolated grains in the made of arfvedsonite-eckermannite ± aegirine granite and groundmass or inclusions in plagioclase phenocrysts, and syenodiorite and quartz-syenodiorite. Syenodiorites of the are occasionally chloritized. In the groundmass of porphy- outer margin include several alkali feldspar and quartz ritic basalts, small grains of clinopyroxene fill the intersti- crystal basement xenoliths. tial space between plagioclase laths. The volcanic rocks include lava flows and pyroclastic Fe–Ti oxide microphenocrysts (titanomagnetite and rocks exposed in the center and north of the complex. ilmenite) represent 0.4 ± 0.2 vol.%. Extremely fine granules The lava flows are porphyritic and subaphyric plagioclase- and fretted opaque crystals occur in the groundmass. basalts, olivine-basalts, olivine-clinopyroxene-basalts, Alkali feldspar-quartz-epidote-calcite-amphibole can be andesitic basalts, hawaiite, benmoreite, rhyolite, arfvedso- intergrown in clots of secondary origin mostly occurring T nite ± aegirine rhyolite, and unusual high Fe2O3 –Al2O3– in rare amygdaloidal plagioclase-basalts. TiO2 rocks resembling lava flows associated with plagioclase-basalts. The plagioclase-basalt is the volumetri- 3.2. Olivine-basalts and olivine-clinopyroxene-basalts cally most abundant lava type followed by benmoreite, andesitic basalt and felsic lavas. Small preserved exposures Olivine-basalts and olivine-clinopyroxene-basalts are of olivine-basalts and olivine-clinopyroxene-basalts are the porphyritic and consist of phenocryts of olivine in basanite youngest lava flows of the complex (Fig. 1). Pyroclastic and olivine-basalt, olivine and clinopyroxene in hawaiite deposits are chiefly tuff breccias of intermediate to rhyolitic from the northern part, and clinopyroxene alone in hawai- composition, ignimbrite and basaltic lapilli. ite from the southern part of the complex. The groundmass G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332 321 is intersertal to intergranular and locally hyalopilitic. Inter- 5. Clinopyroxene chemistry granular groundmass consists of olivine, clinopyroxene, plagioclase, Fe–Ti oxides, and occasionally apatite needles, Clinopyroxenes were analyzed by electron probe micro- secondary carbonate, titanite and zeolites. analyzis at the Venture Business Laboratory of Kobe Uni- Olivine occurs as euhedral, subhedral or skeletal crys- versity using a JOEL X-8900 electron microprobe equipped tals. Olivine phenocrysts (0.4–2 mm) and microphenocrysts with a wavelength dispersive analytical system. Operating represent about 15–6.5 vol.% in basanite and 3 vol.% in conditions were 15 kV and 12 nA using a focused beam. olivine-basalt. Olivine is <0.1 mm across in the ground- Standards were Si: SiO2, Al: Al2O3, Na: NaCl, Mg: mass and converted into iddingsite, serpentine and chlorite MgO, Ti: TiO2, Fe: Fe2O3, Mn: MnO, Ca: CaSiO3. Correc- and iron oxides. tions were made using atomic number, absorption and Clinopyroxene consists of diopside and rare augite. It fluorescence incorporated routine methods. forms about 9 vol.% of the rock among which 6 vol.% The data confirm the microscopic analysis that clinopy- are phenocrysts. Some of them have small inclusions of roxenes are chiefly diopside and augite (Morimoto et al., plagioclase and olivine. Fe–Al-spinel inclusions occur in 1988) in the alkali basalts, and augite in the plagioclase- clinopyroxene of hawaiite from the south. The largest crys- basalts. Diopside and augite are highly variable in tals show extensive internal melting outlined by a sieve tex- terms of Ca, Ti, Al and to some extent Cr contents (Table ture. Glomerophyric association of augite-plagioclase is 2). sparse in all samples. Concentric and sector zoning and For example, the Ti/Al ratio varies widely even in clin- twinning are common in clinopyroxene. Intergrowth of opyroxenes from the same lava: 1.8–0.2 in olivine-basalts, acicular clinopyroxene and amphibole replace primary 1–0.2 in olivine-clinopyroxene-basalts. In contrast, the clinopyroxene in olivine-basalt. Less commonly, small composition appears fairly constant in porphyritic and randomly oriented clinopyroxene prisms occur in plagio- aphyric plagioclase-basalts. clase in basanite. In the groundmass, clinopyroxene forms Clinopyroxene compositions are in the range rounded or rectangular grains. Plagioclase microlite com- Wo5143Ens36.742.4Fs12.815.5 in olivine-basalts and olivine- positions vary between bytownite and labradorite. Biotite clinopyroxene-basalts, and Wo4736Ens3539.5Fs15.524.5 in with yellow to brownish pleochroism occurs in the ground- plagioclase-basalts (Fig. 2). mass of hawaiite from the south. Ti and Al are lower in clinopyroxene from plagioclase- Titanomagnetite (occasionally converted to titanite) basalts than those from olivine-basalts and olivine-clinopy- coexists with ilmenite. Quartz xenocrysts with wavy roxene-basalts (Table 2). extinction jacketed by clinopyroxene corona, and alkali This difference is of petrological significance since Ti feldspar with disequilibrium texture occasionally occur in and Al contents of clinopyroxenes are related to the crys- basanite. tallization condition and the initial magma composition (Lundstrom et al., 1998; Hill et al., 2000; Wood et al., 4. Radiometric age 2001). Clinopyroxenes of basalts from the Bana complex can be distinguished and classified as titaniferous calcic- Plagioclase was manually and magnetically separated clinopyroxene and weakly non-titanoferous calcic-clinopy- from fresh plagioclase-phyric basalt. Selected fractions roxene based on the amount of Ti and Ca entering the were briefly treated in HF (5%) to remove any glass inclu- crystal. Titaniferous augites occur in olivine-basalts and sions and rinsed in distilled water for accurate and precise olivine-clinopyroxene-basalts whereas non-titaniferous K/Ar determinations at the Research Institute of Natural augites typify plagioclase-basalts. Sciences, Okayama University of Science in Japan. Gener- ally, fresh plagioclases of these basalts have high concen- 6. Whole-rock geochemistry tration of K2O 0.3–1 wt% (microprobe analyses) which is sufficient to perform K/Ar radiometric dating. Major and trace elements were determined using X-ray Plagioclase is assumed to be a closed-system with respect fluorescence spectrometry (Rigaku RIX 3000) at the Center to radiogenic 40Ar and 40K. The results show an average for Instrumental Analysis at Yamaguchi University in radiometric age of 30.1 ± 1.2 Ma (Table 1). Japan. Beads were prepared using 0.9 g powder sample, 4.5 g lithium tetraborate (Li2B4O7) flux and 0.54 g Lithium Iodide. Six GSJ (Geological Survey of Japan) geochemical Table 1 standards were analyzed to verify the accuracy of the anal- Analytical results of the K–Ar dating of transitional tholeiitic basalt of the yses. The results are within the range of recommended val- Bana complex ues of the standards. The estimated accuracy X2 0.0022 is Sample No. Potassium Rad. 40Ar K–Ar age Non-rad. within accepted limit (<0.10–0.15) for major elements. (wt%) (10-8 cc STP/g) (Ma) 40Ar(%) A subset of olivine-basalt (G6, K14), olivine-clinopyrox- K48-B 0.198 ± 0.10 24.35 ± 0.55 31.4 ± 1.7 34.6 ene-basalt (K68), and three plagioclase-basalts (K56, K94, 22.34 ± 0.51 28.8 ± 1.6 36.2 KG3) were selected for ICP-MS analyses of REE at ACT- 30.1 ± 1.2 35.4 Lab in Canada. The results are presented in Tables 3 and 4. 322

Table 2 Selected electronmicroprobe analyses of clinopyroxenes wt% K62-Bs K62-Bs K62-Bs G5-Hw G5-Hw G5-Hw G5-Hw G5-Hw G6-Bn G6-Bn G6-Bn G6-Bn G6-Bn G6-Bn K14-Bn K14-Bn K14-Bn K14-Bn K14-Bn ph micro micro Phcore Phrim Phcore gm gm phcore phcore phrim phrim gm phcore phcore phrim phrcore phrim gm .Keooe l ora fArcnErhSine 5(06 318–332 (2006) 45 Sciences Earth African of Journal / al. et Kuepouo G. SiO2 51.40 49.30 49.74 50.24 51.38 50.51 49.27 51.22 47.78 47.37 46.71 46.90 50.91 49.32 45.93 47.40 50.10 48.82 48.33 TiO2 1.69 2.73 2.30 1.71 0.83 1.01 2.21 1.65 2.21 2.42 2.50 2.45 1.14 2.80 3.59 3.24 2.23 2.46 2.91 A12O3 2.31 3.70 3.40 4.52 5.69 7.20 4.98 2.98 7.50 8.29 8.46 8.47 4.94 5.15 7.05 6.48 4.14 5.23 5.65 Cr2O3 0.00 0.00 0.00 0.16 0.03 0.12 0.07 0.03 0.02 0.00 0.03 0.00 0.50 0.04 0.05 0.13 0.19 0.25 0.02 FeO 8.73 8.75 9.56 8.50 9.39 9.25 8.32 7.62 7.86 8.34 7.30 8.10 5.21 7.64 7.92 5.11 5.69 6.05 6.49 MnO 0.28 0.22 0.20 0.16 0.10 0.17 0.18 0.20 0.19 0.16 0.16 0.18 0.09 0.16 0.11 0.08 0.12 0.08 0.07 MgO 14.47 12.77 13.74 15.54 12.40 12.33 13.88 14.69 12.63 12.26 12.05 12.04 15.05 13.47 12.52 12.88 14.58 13.89 13.40 CaO 21.48 21.94 20.73 19.44 19.46 19.24 19.96 21.22 21.26 21.15 21.42 21.30 21.58 22.24 22.30 22.65 22.69 22.66 22.28 Na2O 0.46 0.80 0.60 0.45 1.24 1.16 0.48 0.40 0.77 0.92 0.85 0.82 0.59 0.45 0.57 0.43 0.34 0.42 0.41 Total 100.83 100.20 100.27 100.73 100.51 101.01 99.36 100.02 100.23 100.93 99.48 100.25 100.01 101.26 100.04 98.41 100.08 99.86 99.60

6 O per formula Si 1.90 1.85 1.86 1.85 1.90 1.85 1.84 1.90 1.78 1.75 1.75 1.75 1.87 1.82 1.73 1.78 1.85 1.81 1.80 Al 0.10 0.16 0.15 0.20 0.25 0.31 0.22 0.13 0.33 0.36 0.37 0.37 0.21 0.22 0.31 0.29 0.18 0.23 0.25 Ti 0.05 0.08 0.06 0.05 0.02 0.03 0.06 0.05 0.06 0.07 0.07 0.07 0.03 0.08 0.10 0.09 0.06 0.07 0.08 Cr 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.00 0.00 0.00 Fet 0.27 0.27 0.30 0.26 0.29 0.28 0.26 0.24 0.24 0.26 0.23 0.25 0.16 0.24 0.25 0.16 0.18 0.19 0.20 Mn 0.01 0.01 0.01 0.00 0.00 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Mg 0.80 0.71 0.77 0.85 0.68 0.67 0.77 0.81 0.70 0.68 0.67 0.67 0.82 0.74 0.70 0.72 0.80 0.77 0.75 Ca 0.85 0.88 0.83 0.77 0.77 0.76 0.80 0.84 0.85 0.84 0.86 0.85 0.85 0.88 0.90 0.91 0.90 0.90 0.89 Na 0.03 0.06 0.04 0.03 0.09 0.08 0.03 0.03 0.06 0.07 0.06 0.06 0.04 0.03 0.04 0.03 0.02 0.03 0.03 Total 4.02 4.02 4.02 4.02 4.00 4.00 4.00 4.00 4.02 4.03 4.02 4.03 4.00 4.01 4.03 3.99 4.00 4.01 4.00 Wo 44.15 46.97 43.67 40.65 44.10 43.98 43.47 44.41 47.13 47.16 48.66 47.84 46.25 47.26 48.49 50.76 47.75 48.45 48.38 En 41.39 38.03 40.29 45.21 39.11 39.20 42.06 42.80 38.94 38.04 38.10 37.64 44.89 39.82 37.89 40.16 42.70 41.31 40.49 Fs 14.46 15.00 16.05 14.13 16.79 16.82 14.46 12.79 13.93 14.80 13.24 14.52 8.86 12.93 13.62 9.08 9.56 10.24 11.12 K91*-l K91*-2 K91*-3 K91*-4 KG3-1 KG3-2 K61-1 K61-2 K61-3 K61-4 K61-5 K61-6 K61-7 K66-1 K66-2 K95-1 K95-2 K95-3 K74-1 K74-2

SiO2 49.94 46.43 49.44 47.98 50.55 51.46 49.54 49.56 50.33 48.98 49.08 49.71 51.31 51.15 52.47 52.88 51.77 50.60 53.17 51.29 TiO2 1.31 3.06 1.31 2.29 1.11 0.66 1.53 1.57 1.22 1.64 1.87 1.67 1.28 0.46 0.19 0.20 1.09 1.44 0.22 0.96 A12O3 2.55 5.09 2.93 4.26 1.95 1.40 3.11 3.13 2.27 3.68 3.50 2.98 1.78 1.98 0.84 0.67 2.06 2.11 4.36 1.55 Cr2O3 0.00 0.01 0.00 0.00 0.00 0.00 0.05 0.02 0.04 0.00 0.00 0.01 0.01 0.01 0.01 0.00 0.00 0.00 FeO 12.69 11.70 13.73 12.56 13.09 12.52 9.97 10.12 9.84 10.65 10.25 10.89 12.61 10.60 8.60 7.39 9.58 13.90 14.12 11.92 MnO 0.60 0.38 0.50 0.45 0.35 0.32 0.23 0.31 0.34 0.33 0.34 0.36 0.44 0.57 0.65 0.77 0.30 0.43 0.27 0.28 MgO 12.05 11.18 11.75 11.37 14.40 11.67 13.92 14.61 14.80 13.67 14.03 14.20 13.85 13.71 13.74 14.72 14.83 12.19 9.76 14.82 CaO 19.83 20.49 19.59 20.38 17.42 21.38 20.42 19.73 20.19 20.23 19.57 19.69 17.95 20.03 23.19 23.14 20.09 19.06 15.84 18.27 Na2O 0.45 0.70 0.40 0.43 0.32 0.30 0.46 0.68 0.42 0.41 0.45 0.55 0.03 0.35 0.21 0.06 0.32 0.19 1.67 0.00 Total 99.45 99.06 99.66 99.73 99.21 99.71 99.23 99.72 99.47 99.59 99.09 100.09 99.26 99.01 99.91 99.84 100.05 99.93 99.41 99.08 G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332 323

Fig. 2. Clinopyroxene compositions of alkali basalts (open circle) and transitional tholeiitic basalts from Bana (shaded ﬁeld) complex obtained from plots in comparison with clinopyroxene from Ethiopian transitional and tholeiitic basalts. Symbols: diamond, clinopyroxene in alkali basalts; ﬁlled circles, clinopyroxene in transitional tholeiitic basalts. Abbreviations: AB, alkali basalt; TAB, transitional alkali basalt; TTB, transitional tholeiitic basalt; TB, tholeiitic basalt; S line, Skaeggard pyroxene trend.

6.1. Chemical alteration

The effects of alteration were minimized in the first instance by careful selection of freshest samples. The most sensitive elements to weathering are alkali elements (Na, K, Rb) as their concentration is affected even in young and fresh basalts. This mobility is influenced by breakdown of interstitial glass and alkali bearing phases, and associated development of secondary minerals. The mobility of Na and K is manifested by the scatter in SiO2 vs. K2O+Na2O plot (Fig. 3). Furthermore, LIO up to 4 wt% in a few samples indicate alteration. However, the overall coherent variation of most major and trace elements suggests that alteration was limited to highly mobile alkali elements. Therefore, the use of mobile

6 O per formula SiAlTiCr 1.91Fet 0.11Mn 1.79 0.04Mg 0.23 0.00Ca 0.41 1.89 0.09Na 0.02 0.13 0.00 0.38 0.69Total 1.83 0.04 0.01 0.19 0.00 0.81Wo 0.44 0.64 1.92 0.07 0.03En 0.02 4.01 0.09 0.00 0.85Fs 0.40 0.67 1.96 0.03 0.05 42.21 0.01 4.03 0.06 0.00 0.80 1.87 0.42 0.65 35.70 0.02 0.03 45.06 0.14 0.01 4.02 0.00 0.83 22.09 34.19 0.04 0.40 0.82 41.63 1.87 0.03 0.00 0.01 20.74 4.02 0.14 0.71 34.75 0.32 43.97 0.66 1.90 0.04 0.02 0.01 23.62 34.12 4.02 0.10 0.00 36.33 0.87 0.79 0.32 1.85 21.91 0.03 0.02 41.79 44.86 0.16 0.83 0.01 4.00 0.00 1.86 21.89 0.03 42.76 0.31 0.05 0.82 34.09 0.16 4.03 0.01 0.00 40.56 21.04 0.80 0.05 1.87 0.34 40.92 0.83 0.00 0.05 0.13 16.67 0.01in 0.32 4.04 1.94 42.18 0.82 0.05 41.43 0.77 0.01 0.08conjunction 0.03 0.00 16.89 42.26 42.29 1.94 4.03 0.82 0.79 0.34 0.04 41.31 0.09 16.31 0.03 0.00 0.01 39.78 0.79 1.97 4.03 0.40 0.01 41.23 0.80 17.93 0.03 40.82 0.04 0.00 0.01 4.03 17.46 1.97 0.79 0.34 0.01 37.87 40.97 0.78 0.03 0.04 0.02 0.00 18.22 41.87 40.64 with 0.73 4.04 0.27 0.01 0.78 1.93 0.00 21.49 0.00 0.02 39.89 46.81 0.09 0.81 3.98 0.23 0.77 18.23 1.92 46.28immobile 38.60 0.03 0.03 0.02 0.09 4.01 0.93 0.00 14.58 40.97 0.82 0.30 0.04 0.01 2.00 41.48 12.76 0.92 4.02 0.00 0.01 0.19 40.36 42.61 0.44 0.00 0.82 0.01 1.94 4.01 15.91 0.01 35.94major 0.00 0.07 38.96 0.80 0.69 0.44 23.70 0.03 0.02 33.41 0.01 37.73 0.78 4.01 0.00 27.62 0.55 0.38 (e.g. 0.01 42.59 4.00 0.01 0.64 19.67 0.84 0.12 Ti) 3.96 0.74 0.00 and 4.00 trace 324 G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332

Table 3 6.2. Major element compositions XRF whole-rocks analyses of Bana alkali basalts wt% Bn K14 Bn G6 B K62 Hw G5 Hw K68 Major element compositions (Fig. 4) show a great vari-

SiO2 44.42 44.36 46.36 48.46 48.73 ation of MgO and similar TiO2 contents (2.1–3.6 wt%) at TiO2 3.16 3.31 3.25 2.79 2.55 small differences in SiO2 content between olivine-basalts Al2O3 14.59 14.08 15.44 14.56 15.34 and olivine-clinopyroxene-basalts; and plagioclase-basalts. Fe2O3 11.86 13.14 12.99 11.95 11.69 Olivine-basalts are characterized by 8.5–9.5 wt% MgO MnO 0.16 0.20 0.17 0.17 0.17 MgO 9.34 8.74 6.84 7.44 6.59 and 44–44.5 wt% SiO2 and a total range of Mg# 62– CaO 9.59 10.49 8.48 7.94 7.35 66.1, olivine-clinopyroxene-basalts have 6.6–7.5 wt% MgO Na2O 3.17 3.01 3.25 3.69 3.70 and 46.3–48.7 wt% SiO2 and a total range of Mg# K2O 1.48 0.94 1.17 1.57 1.53 56.6–63.8; and plagioclase-basalts have 2.2–5.2 wt% MgO P2O5 0.72 0.60 0.53 0.72 0.74 and 47.3–51.1 wt% SiO2 and a total range of Mg# 36– Total 98.49 98.88 98.47 99.27 98.37 47.7 (Tables 3 and 4). Among the plagioclase-basalts, the Mg# 66.11 62.22 56.60 63.80 61.49 subaphyric and weakly porphyritic samples (K91, K74, A.I 0.47 0.43 0.50 0.42 0.53 K64 and K56) show the highest MgO contents 5.2– Norm qz 4.5 wt% relative to the highly porphyritic samples at similar ab 16.57 17.95 27.91 31.43 31.78 silica contents. Following the classification grid of Le Bas an 21.54 22.36 24.48 18.67 21.11 et al. (1986) (Fig. 3), olivine-basalts are basanites, olivine- or 8.88 5.63 7.01 9.34 9.16 clinopyroxene-basalts are basalts and hawaiites, and pla- di 17.68 21.26 11.95 13.07 8.90 hy 2.02 2.63 9.05 gioclase-basalts are basalts and rare hawaiites. Further mt 3.49 3.85 3.82 5.23 5.17 division (Irvine and Baragar, 1971) allows distinction of ilm 6.10 6.36 6.27 5.35 4.93 alkaline and subalkaline magma series in the Bana com- ol 17.32 15.88 14.23 11.78 7.34 plex. The olivine-basalts and olivine-clinopyroxene-basalts ne 5.76 4.24 are alkaline whereas plagioclase-basalts are subalkaline ap 1.73 1.43 1.28 1.71 1.77 (Fig. 3). ppm The distinction between the Bana alkali and subalkalic Ba 605 422 499 465 592 basalts also appears on the Ne–Ol–Di–Hy–Q normative Rb 37 23 31 34 29 Th 2 3 9 8 3 system (Fig. 5). Normative minerals were calculated using Sr 954 759 843 659 861 the adjusted Fe2O3/FeO = 0.2 (Middlemost, 1989) for the Nb 74 52 67 68 71 two groups. The compositions range from quartz-tholeiite Ta 4 5 7 basalts, through olivine- to nepheline-bearing alkali bas- Zr 245 190 234 264 278 alts. The absence of calcium-poor pyroxene in subalkalic Y2727233231 Co 58 54 68 60 52 basalts, an essential and critical feature of tholeiites, pro- Cr 273 117 152 334 166 hibits their classification as true oversaturated tholeiites Cu 32 35 30 56 40 and suggests instead that they are transitional tholeiitic Ga 18 20 17 19 19 basalts, in opposition to transitional alkalic basalts follow- Ni 162 80 106 134 119 ing the terminology by Tilley (1950) and Coombs (1963). Pb 1 0 0 3 5 V 225 207 183 282 183 Basalts having similar characteristics occur in the Turkana Zn 111 105 111 98 111 locality in northern Kenya (Bellieni et al., 1981) and in Aiba in Ethiopia (Zanettin et al., 1980). The field (Fig. 4) REE La 43.1 44.9 48.4 for the Bana transitional tholeiitic basalts is mainly within Ce 81.1 87.2 90.0 the 1 atm cotectic, indicating equilibration in a very shal- Pr 9.80 10.6 10.9 low magma reservoir. The common features between alkali Nd 40.8 44.7 43.9 and transitional tholeiitic basalts in the Bana complex are Sm 9.05 9.89 8.75 the high titanium ( 2.2–3.5 wt% TiO ) character, similar Eu 2.85 2.93 2.70 2 Gd 7.38 8.38 6.90 CaO/Na2O ratios (2–3.5) and the high potassium con- Tb 0.97 1.17 0.86 tents (0.9–1.6 wt% in alkali basalts; 0.6–1.8 wt% in transi- Dy 5.00 6.51 4.56 tional tholeiitic basalts). Five samples of transitional Ho 0.87 1.19 0.76 tholeiitic basalts (G1, G9a, K93, K95 and K105) have high Er 2.15 3.11 1.92 Al O contents (18.36–20.35 wt%). Tm 0.281 0.440 0.252 2 3 Yb 1.72 2.66 1.54 Lu 0.225 0.358 0.203 6.3. Trace element compositions

Relative to transitional tholeiitic basalts, the alkali bas- elements (Zr, Nb, Y and HREE) of these rocks for petro- alts have twice or more Nb contents, and higher Ni, and to genetic interpretations is a valid approach (e.g. Pearce a certain degree Sr contents (Fig. 4). Other trace elements and Cann, 1973; Winchester and Floyd, 1976). notably the most incompatible ones such as Ba and Rb G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332 325

Table 4 XRF whole-rocks analyses of Bana transitional tholeiitic basalts (TrTB) wt% TrTB K91 K64 G9a G1 K95 K104 K54 K93 K67 K74 K58 K77 K39’ K48-b K56 KG3 K107

SiO2 47.36 45.56 49.68 49.36 49.29 47.12 49.06 49.56 49.43 48.99 49.13 48.54 51.16 48.50 48.94 49.66 48.67 TiO2 3.11 3.29 2.71 2.15 2.70 2.66 3.78 2.82 3.10 3.37 3.87 3.43 3.10 3.52 3.38 3.52 3.40 Al2O3 13.31 14.50 19.14 20.35 19.39 19.10 14.60 18.36 15.28 14.85 14.52 15.60 15.23 15.97 15.20 15.14 14.35 Fe2O3 14.18 14.23 10.63 9.86 10.53 10.11 14.75 11.33 13.24 14.11 13.70 13.18 12.34 13.72 14.19 13.72 13.04 MnO 0.26 0.18 0.13 0.11 0.14 0.12 0.20 0.14 0.21 0.18 0.19 0.18 0.19 0.17 0.16 0.17 0.18 MgO 4.32 5.25 2.76 2.24 2.85 2.68 3.84 2.89 3.74 4.70 3.70 3.78 3.27 3.96 4.49 4.19 3.03 CaO 8.29 9.61 9.82 10.44 10.15 9.92 8.18 9.80 7.21 7.34 7.55 9.07 8.42 9.18 8.72 7.48 8.46

Na2O 2.70 2.20 3.35 3.34 3.12 3.27 2.76 3.24 3.49 3.07 3.27 3.12 3.12 3.33 3.11 3.45 2.96 K2O 1.40 0.89 0.78 0.55 0.80 0.62 1.05 0.65 1.80 1.00 1.09 1.18 1.67 0.92 1.08 1.27 1.08 P2O5 1.97 0.39 0.32 0.31 0.35 0.28 0.56 0.36 0.94 0.52 0.55 0.45 0.60 0.44 0.46 0.56 0.57 Total 96.90 96.09 99.30 98.70 99.32 95.89 98.78 99.13 98.43 98.13 97.57 98.54 99.09 99.71 99.72 99.14 95.73 Ms# 43.00 47.75 39.15 36.02 40.08 39.66 39.19 38.69 41.15 45.22 40.10 41.55 39.63 41.69 43.92 43.05 36.55 A.I 0.45 0.32 0.33 0.30 0.31 0.32 0.39 0.33 0.50 0.41 0.45 0.41 0.46 0.41 0.41 0.46 0.42 Norm qz 4.21 0.81 1.88 1.73 1.95 0.39 5.04 2.73 1.01 3.05 4.07 1.26 4.44 0.17 1.07 1.83 5.43 ab 23.57 19.37 28.49 28.59 26.57 28.88 23.62 27.66 30.01 26.43 28.32 26.75 26.60 28.27 26.35 29.40 26.12 an 20.71 28.14 35.16 39.45 36.78 37.12 24.67 33.92 21.04 24.24 22.28 25.46 22.85 25.97 24.42 22.30 23.71 or 8.56 5.48 4.61 3.27 4.78 3.83 6.26 3.88 10.80 6.04 6.60 7.09 9.94 5.47 6.37 7.54 6.67 di 7.01 15.49 9.83 9.18 9.67 10.19 10.55 10.62 7.53 7.80 10.32 14.30 12.75 13.95 13.18 9.48 13.58 hy 19.72 17.78 10.15 9.22 10.36 9.74 15.74 10.73 16.44 19.38 14.38 12.52 11.47 13.35 15.84 16.29 11.34 mt 4.24 4.29 3.10 2.90 3.08 3.06 4.33 3.31 3.90 4.17 4.07 3.88 3.61 3.99 4.13 4.01 3.95 ilm 6.10 6.50 5.18 4.13 5.16 5.27 7.27 5.40 5.99 6.52 7.53 6.61 5.93 6.71 6.43 6.74 6.74 ol ne ap 4.81 0.96 0.76 0.74 0.83 0.69 1.35 0.85 2.25 1.24. 1.33 1.07 1.44 1.04 1.09 1.33 1.40 ppm Ba 798 258 282 256 261 222 467 265 517 507 409 320 509 361 327 540 451 Rb 58 22 13 12 18 13 19 10 38 24 20 32 42 20 17 36 25 Th155646695114 66546316 Sr 551 551 584 632 620 574 451 564 518 512 439 507 514 518 485 480 522 Nb 36 23 24 23 23 21 32 24 44 33 39 33 37 31 32 34 33 Ta 3 32 Zr 254 168 174 168 168 154 265 164 340 266 282 230 278 223 229 287 244 Y 4727232426233626514144364032354237 Co 31 55 64 43 44 33 58 48 43 52 43 43 46 46 51 46 41 Cr 6 67 8 24 7 13 15 8 6 29 6 16 14 11 22 19 9 Cu nd nd 26 48 12 15 21 28 nd 7 18 88 nd 69 27 17 13 Ga 19 19 23 22 24 23 24 24 28 24 24 22 24 22 23 23 23 Ni nd 53 16 23 6 13 17 8 nd 33 7 16 10 11 20 24 8 Pb 2 0 0 nd nd nd 4 0 2 2 0 nd 3 1 nd 1 1 V 129 324 218 178 210 231 343 221 158 222 286 262 203 309 285 226 258 Zn 161 116 96 91 99 87 137 97 175 146 147 115 137 123 130 139 131 REE La 18.97 24.23 32.23 Ce 39.40 50.80 66.70 Pr 5.30 6.70 9.10 Nd 24.00 30.70 40.40 Sm 6.04 7.94 9.76 Eu 2.05 2.62 3.12 Gd 5.85 7.60 9.58 Tb 0.85 1.14 1.36 Dy 4.77 6.41 7.74 Ho 0.87 1.19 1.40 Er 2.25 3.06 3.70 Tm 0.32 0.44 0.53 Yb 1.86 2.51 3.04 Lu 0.25 0.34 0.41 have similar abundances in transitional tholeiitic basalts rocks are characterized by a narrow range of Rb/Sr ratio than and alkali basalts (Fig. 4). The two groups of basaltic 0.03–0.04, except for one sample (K91) of transitional 326 G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332

12 tholeiitic basalts with a ratio of 0.08. This narrow ratio Rhyolite range is within the limit of mantle peridotites (Rb/ 10 Tr Sr < 0.07). O)

2 Be The Y/Nb and Zr/Nb ratios (Fig. 6a) are consistent dis- 8 Alkalin e criminants between the two groups of basalts: alkali basalts

O+K Mu 2 6 Hw have low ratios of Y/Nb < 1 and Zr/Nb < 4 while the tran- Bn sitional tholeiitic basalts are characterized by their high 4 ratios of Y/Nb > 1 and Zr/Nb > 6.

wt%( Na Ba Subalkaline The Zr/Nb–Zr/Y and Zr–Nb diagrams (Weaver et al., 2 1972; Menzies and Kyle, 1990) discriminate between alkali 0 basalts and transitional tholeiitic basalts of the Bana com- 40 45 50 55 60 65 70 75 80 plex (Fig. 6a–c).

wt% SiO2 The Primary Mantle-normalized (Sun and McDonough, 1989) trace element patterns (Fig. 7a and b) have negative Fig. 3. Total Alkali – Silica classification (Le Bas et al., 1986). Straight Th and K anomalies relative to similar incompatible ele- line in the basaltic field delineates the boundary between alkaline and subalkaline basaltic series after Irvine and Baragar (1971). Symbols: alkali ments (Ba and Ta) in alkali basalts and transitional tholei- basalt group (basanite and basalts (filled diamonds); hawaiites (filled itic basalts. square)); transitional tholeiitic basalts (filled circle). The circled area In spite of these differences, all of these lavas carry the delineates the composition of transitional basalts of the Bamoun Plateau typical OIB signature of Nb–Ta enrichment (Fig. 7a and b). (Moundi, 2003).

10 209 8 159 6 109 4 Ni ppm MgO wt% MgO 2 59

0 9

5 1000

4 800 600 wt%

2 3 400 Ba ppm TiO 2 200

1 0

16 15 70 14

wt% 13 50 3

O 12 2 Sr ppm

Fe 11 30 10 9 10

22 40 20 18 30

wt% 16 3

O 14 2

Rb ppm 20

Al 12 10 8 10 44 46 48 50 52 44 46 48 50 52 SiO2 wt% SiO2 wt%

Fig. 4. Variation diagrams of majors and trace elements vs SiO2 for transitional tholeiitic basalts and alkali basalts from the Bana complex. Symbols are from Fig. 3. G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332 327

Di 12 Ne Qz Low degree PM Parana low-Ti lavas 10

1 atm 8 Zr/Y

6 High degree PM Afar Riftt transitional basalts 4 0246810 (a) Zr/Nb Ol Hy

Fig. 5. Ne–Ol–Di–Hy–Q normative system showing the plot of the 80 plagioclase-basalts from Bana in the field of quartz-tholeiite. Oval field 70 delineates the fields for transitional tholeiitic basalt from Afar rift zone (in 60 Northern Turkana, Kenya by Bellieni et al. (1981); and in Aiba, Ethiopia by Zanettin et al. (1980)) and Dash oval line delineates the fields for 50 Parana-Etendeka low-Ti lavas (Bellieni et al., 1984). Cotectic at 1 atm for 40 equilibrium Ol–Pl–Cpx-basaltic liquid is from Thompson (1982); arrows Nb ppm 30 indicate the direction of falling temperature. Abbreviations: Ne, Nephe- 20 line; Ol, Olivine; Di, Diopside; Hy, Hypersthene; Q, Quartz. Symbols are from Fig. 3. 10 0 0 100 200 300 400 (b) Zr ppm The rare earth element (REE) composition (Fig. 8a and b) provides a good distinction and reveals the intrinsic transitional tholeiitic character of these basalts. 300 There is an increase of light vs heavy REE enrichment 250 from transitional tholeiitic basalt to alkali basalts: the 200 [La/Yb]N ratios = 5.37–5.89 in transitional tholeiitic bas- Zr ppm alts and 9.37–17.48 in alkali basalts; a relatively flat heavy 150 rare earth element (HREE) curve ([Tb/Yb]N = 1.81–2.32) 100 in both basaltic types portraying the general array of 0.20 0.40 0.60 0.80 1.00 1.20 1.40 chondrite-normalized (Anders and Grevesse, 1989) pat- Y/Nb Transitional (c) Alkali basalts terns of Afar Rift transitional basalts (Barberi et al., basalts 1975)(Fig. 8b). The [Tb/Yb]N ratios are comparable with Fig. 6. (a) Zr/Y against Zr/Nb plot showing qualitative trend of those of alkali basalts from Hawaii [Tb/Yb]N = 1.89–2.45 compositions resulting from low vs high melting degrees of the Bana which are commonly considered to have been generated basaltic rocks. Also note the depleted character of transitional tholeiitic in a garnet-bearing lherzolitic mantle (McKenzie and basalts compared with alkali basalts. Abbreviation: PM, partial melting. O’Nions, 1991, 1995; Frey et al., 1991). (b) Nb against Zr plot showing that olivine-, olivine-clinopyroxene-basalts (alkali basalts) and plagioclase-basalts (transitional tholeiitic basalts) from the Bana complex represent two different magma types as indicates the 7. Discussion presence of two covariant trends. (c) Plot of Zr against Y/Nb ratios distinguishing the basaltic rocks from the Bana complex into alkali basalts From the data summarized above, olivine-basalts and and transitional basalts. The boundary (discontinued line) is from Pearce olivine-clinopyroxene-basalts (i.e. alkali basalts) on the and Cann (1973). Symbols are from Fig. 3. one hand, and plagioclase-basalts on the other hand (i.e. transitional tholeiitic basalts) represent basic magmas clo- neglected (Barberi et al., 1971). In the Bana plagioclase- sely associated in space. The plagioclase-basalts are more basalts, the early crystallization and withdrawal of clinopy- evolved magmatic products (Mg# 36–47.7) than the for- roxene phenocrysts out of the system followed by the mer (Mg# 46.3–66.1). In contrast, primitive lava from abundant crystallization of calcic plagioclase phenocrysts, the upper mantle has a Mg# of 68–72 (Green, 1971). The may have led to the crystallization of groundmass clinopy- occurrences of titaniferous augites in olivine-basalts and roxene with low Ti content. Such a sequence of crystalliza- olivine-clinopyroxene-basalts, and the predominance of tion has been reported in the Erta’Ale plagioclase non-titaniferous augites in plagioclase-basalts reflect the porphyritic rocks of mildly alkaline series (Barberi et al., silica-undersaturated character of the former. The compo- 1971). Progressive Ca depletion and Fe enrichment of clin- sition of clinopyroxene of undersatured basaltic magmas is opyroxenes in passing from alkali to transitional tholeiitic considered to be highly sensitive to the order of appearance basalts in Bana may suggest a genetic relationship at first of clinopyroxene-calcic plagioclase if pressure effects are glance. 328 G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332

1000 The transitional tholeiitic basalts were not derived from the exposed alkali basaltic melt by fractionation of olivine Alkali basalts and/or clinopyroxene, because of their low concentrations 100 in LILE, Nb, Zr and Y compared with those of alkali basalts. Other lines of evidence such as the high silica and low 10 magnesia contents, La/Yb vs. Dy/Yb and La/Sm vs. Sm/ Yb diagrams (Fig. 9a and b) suggest that the parental melt

Rocks/Primitive Mantle of transitional tholeiitic basalts were probably formed by high degree partial melting of the mantle at shallow depth, 1 (a) Rb Ba Th K Ta Nb La Ce Sr NdP Hf Zr Sm Ti Tb Y Tm Yb compared to alkali basalts (Thirlwall et al., 1994; Bogaard and Worner, 2003). Geophysical data (Fairhead and 1000 Okereke, 1987; Plomerova et al., 1993; Poudjom Djomani et al., 1995) give supporting evidence of crustal thinning Transitional tholeiitic basalts beneath the Cameroon Line. Thus occurrence of basalts 100 deriving from partial melting of the mantle at shallow depth in these settings seems probable. The distinctive high alumina contents (Al2O3: 18.36– 10 20.35 wt%) in the highly porphyritic transitional tholeiitic Rocks/Primitive Mantle 1 (b) Rb Ba Th K Ta Nb La Ce Sr NdP Hf Zr Sm Ti Tb Y Tm Yb 4.0 Garnet peridotite Fig. 7. Spider diagram using the Primary Mantle normalization values 3.5 8% recommended by Sun and McDonough (1989) for: (a) Olivine-, olivine- 10% 3.0 TrTB 12% clinopyroxene-basalts (alkali basalts), and (b) Plagioclase-basalts (transi- 14% K14 G6 tional tholeiitic basalts). Symbols are from Fig. 3. 2.5 16% K68 2.0

Dy/Yb 1.5 15%10% 1% 1.0 1000 Spinel peridotite Alkali basalts 0.5 0.0 0 10203040 100 (a) La/Yb

7 10 G6 Rock/Chondrite 6 1% 5 K68 K14 Melt model 1 4 (a) La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu 5% 10% La/Sm 3 15% TrTB 1000 2 1% Transitionalt holeiitic basalts 5% 1 15% 10% MORB-source 100 0 0102468 (b) Sm/Yb

10 Fig. 9. (a) La/Yb vs Dy/Yb for olivine basalts (G6, basanite; K14,

Rock/Chondrite basanite; olivine-clinopyroxene-basalt (K68, hawaiite); and plagioclase- basalts (transitional tholeiitic basalts, TrTB) from the Bana complex. Curves are degrees of melting for garnet peridotite, and spinel peridotite 1 (Thirlwall et al., 1994; Bogaard et al., 2003). The good fit with garnet (b) La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu peridotite as possible source region of basaltic rocks from Bana should be noted as well as higher degree of melting to generate the TrTB. (b) Sm/Yb Fig. 8. Rare earth element distributions diagrams normalized to Chon- vs La/Sm plot gives the melt curve for inverse batch melting model and the drite values recommended by Anders and Grevesse (1989) for: (a) Olivine-, depleted MORB source composition. TrTB apart from higher degree of olivine-clinopyroxene-basalts (alkali basalts), and (b) Plagioclase-basalts melting generated from enriched source composition relative to N-MORB (transitional tholeiitic basalts). Symbols are from Fig. 3. from Sun and McDonough (1989). Symbols are from Fig. 3. G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332 329 basalts (G1, G9a, K93, K95 and K104) contrasting with island basalts (OIB) can generally be either (1) direct par- absence of positive Eu anomalies (Eu/Eu* = 0.96–1.0; tial melting of plumes that have risen from below a thermal Fig. 8a) might reflects the abundance of plagioclase rather boundary layer separating the lower mantle from the con- than its accumulation by gravitational sinking. vecting upper mantle or (2) preferential melting of dis- Tatsumi et al. (1986) showed that Zr and Y are both rel- persed HIMU-OIB streaks or blobs from a digested atively immobile in aqueous fluids and can provide infor- oceanic crust within the convecting upper mantle (e.g. mation on the mantle source region of lavas without Zindler et al., 1984). Model number 2 was discarded by contamination by crustal aqueous fluids. Fig. 6a–c clearly Halliday et al. (1990) for the genesis of Cameroon Line suggest again that alkali basalts and transitional tholeiitic basalts. Hence model number 1 could be considered to basalts from the Bana complex are two different types of explain the above variations between the two groups of basalts rather than one being derived from the other by basalts in Bana whereby the parent magmas of alkali bas- crystal fractionation. Another process is therefore required alts originated from a secondary enriched plume coming to explain the origin of these two groups of basalt. Small from the top of the dome near the depth of the transition and high degrees of partial melting of the mantle at high zone at the location of the African superswell (Courtillot pressure (>15 kbar) could yield high variation of Zr/Y et al., 2003) whereas that of transitional tholeiitic basalts ratios in the resulting melt due to retention, or release of originated in a depleted asthenosphere as evidenced by Y in garnet and clinopyroxene (Tatsumi and Eggins, the high Zr/Nb, low Nb and low [Ce/Yb]N ratios. Condie 1995). This may explain the occurrences of basanite and (2003, 2005) suggests that in modern and Archean basalts, hawaiite (Fig. 6a). it is possible to characterize plume and non-plume mantle The highly incompatible element ratios such as Zr/Nb, sources with four incompatible element ratios: Nb/Th, K/Nb, Ba/Th, Th/La, Ba/La and Ba/Nb are shown to be Zr/Nb, Zr/Y, and Nb/Y. Using these four ratios, alkali least susceptible to fractionation during partial melting and transitional tholeiitic basalts from the Bana complex (Table 5), and are not significantly fractionated during lim- plot above the DNb line in the mantle plume field as ited degrees of low-pressure crystallization of OIB magmas defined by Fitton et al. (1997). (Weaver, 1991), hence they could be useful indicators for It is suggested that these basalts have chemical features basaltic end-member characterization in Bana. These ratios consistent with plume activity outlined by their OIB char- suggest that: (1) the alkali basalts source is influenced by acters (Fig. 10a and b). the HIMU end-member in terms of Zr/Nb and Th/La; The HREE ratios and simple interpretation of Fig. 8 by EM2 end-member in terms of Ba/Th; Ba/La suggests suggest moderate fractionation and a garnet peridotite res- the contribution of EM1; whereas (2) the transitional idue expressed in [Tb/Yb]N = 1.81–2.32 for alkali basalts tholeiitic basalts source is largely influenced by the EM1 and [Tb/Yb]N = 1.84–1.88 for transitional tholeiitic bas- end-member except for the Zr/Nb ratio that suggests alts. This common source region should be located in the contribution of EM2 end-member. spinel-lherzolite and garnet-lherzolite transition zone cor- The overall OIB character of the two groups of basalts responding to depths of about 70–80 km (Frey et al., in accordance with the models for the genesis of oceanic 1991; McKenzie and O’Nions, 1991) in the upper mantle.

Table 5 Trace element ratios of selected olivine-basalts (K14, G6), olivine-clinopyroxene-basalt (K68); and transitional tholeiitic basalts (K56, K95, KG3) from Bana compared with diﬀerent OIB end-menbers, N-MORB, PM, and average continental crust values from Weaver et al. (1991) Sm/Nd Zr/Nb Ba/Th Th/La Ba/La Ba/Nb K/Nb Rb/Sr K14 0.22 3.76 122.49 0.12 14.15 8.30 167.15 0.04 G6 0.22 4.43 91.51 0.12 10.57 6.97 115.12 0.06 K68 0.20 3.98 105.47 0.10 10.76 7.51 182.53 0.04 K56 0.26 7.50 112.55 0.11 11.84 8.75 272.31 0.04 K95 0.25 7.67 125.89 0.10 13.07 10.12 272.41 0.03 KG3 0.24 9.20 187.37 0.09 16.57 15.30 301.12 0.08 Reference values from Weaver (1991) Chondrite 0.33 Upper Cru 0.17 HIMU-mini 2.7 39.0 0.1 6.2 HIMU-max 5.5 85.0 0.2 9.3 EM1-mini 3.5 80.0 0.1 11.3 EM1-max 13.1 204.0 0.2 19.1 EM2-mini 4.4 57.0 0.1 7.3 EM2-max 7.8 105.0 0.2 13.5 N-MORB 30.0 60.0 4.0 4.3 P.M 14.8 77.0 0.1 9.0 Cont. C. av. 16.2 124.0 0.2 54.0 330 G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332

100 (3) the transitional tholeiitic and alkali basalts of the Bana complex represent two spatially close, but tem- ARC DM NMORB porally different magmatic series, PM (4) temporal transition from transitional tholeiitic bas- EN OPB 10 alts to alkali basalts in the Bana system may be DEP ascribed to decreasing degrees of partial melting of Zr/Nb EM2 the mantle coupled with the increase of pressure or OIB EM1 depth of melting through time, REC HIMU (5) low La/Nb, Ba/Nb, and Rb/Sr ratios suggest insignif- 1 01020 icant contamination of these basalts by the crust, (a) Nb/Th however isotopic data are required to confirm this aspect of the work, 10.00 (6) the Cenozoic basalts from the Bana complex are OIB dominantly transitional tholeiitic with subsidiary REC Plume source alkali basalts in contrast to the overall Cameroon EM2H HIMU-EM1 Line basaltic composition, 1.00 OPB (7) given the genetic relationship between basaltic rock series and tectonic setting, the finding of transitional PM Nb/Y tholeiitic basalts in Bana may imply that this area 0.10 DEP represents a zone of maximum extension of the lithosphere below the Cameroon Line in Cenozoic times. Non-Plume source Nb line 0.01 110Acknowledgements (b) Zr /Y The authors wish to thank the administration of the Fig. 10. Plot of basalts from Bana relative to the mantle compositional components (filled star) and fields for basalts from various tectonic Venture Business Laboratory at the Kobe University for settings as defined by Weaver (1991) and Condie (2003). (a) Nb/Th–Zr/ allowing the electron probe microanalyses. We are grateful Nband. (b) Zr/Y–Nb/Y. Abbreviations: PM, primitive mantle; DM, to C.I. Chalokwu, J.M. Bardinzeff and late A.B. Kampu- shallow depleted mantle; HIMU, high mu (U/Pb) source; EM1 and EM2, nzu for their constructive comments and corrections on enriched mantle sources; ARC, arcrelated basalts; N-MORB, normal manuscript. F. Schwandner (University of Arizona) is ocean ridge basalt; OIB, oceanic island basalt; DEP, deep depleted mantle; EN, enriched component; REC, recycled component; OPB, oceanic acknowledged for improving the edition of the early man- plateau basalt. Symbols are from Fig. 3. uscript. The manuscript has been greatly improved thanks to the thorough and constructive comments of reviewers S. Muhongo and an anonymous colleague. The senior author was supported by a fellowship from the Japanese Ministry Differences in LREE/HREE such as [La/Yb]N = 9.37– of Education, Science, Sport and Culture (Monbushu). 17.48 in alkali basalts and [La/Yb]N = 5.37–5.89 in transitional tholeiitic basalts also agree with the variable degrees F.M. Tchoua, E. Njonfang and P. Kamgang of University and depths of partial melting of the mantle source. of Yaounde I (Cameroon) are acknowledged for fruitful The K/Ar radiometric age of 30.1 ± 1.2 Ma of the Bana discussions during the field trip. transitional tholeiitic basalts is younger than those of transitional alkali basalts of Mount Bangou 42–44 Ma (Fosso References et al., 2005) and plateau Bamoun 51 Ma (Moundi, 2004) in the Western Cameroon Highlands. Similar age ranges Anders, E., Grevesse, N., 1989. Abundances of the elements: Meteoritic (44–51 Ma) of transitional basalts of the East African Rift and solar. Geochimica et Cosmochimica Acta 53, 197–214. zone imply that the lithospheric extension was continental- Barberi, F., Bizouard, H., Varet, J., 1971. Nature of clinopyroxene and iron enrichment in alkalic and transitional basaltic magmas. Contri- wide in Africa, at least in the Cenozoic. butions to Mineralogy and Petrology 33, 93–107. Barberi, F., Ferrara, G., Santacroce, R., Treuil, M., Varet, J., 1975. A transitional basalt-pantellerite sequence of fractional crystallization, 8. Conclusion the Boina center (Afar Rift, Ethiopia). Journal of Petrology 16, 22– 56. (1) Plagioclase-basalts of the Bana complex are transi- Bellieni, G., Visentin, E.J., Piccirillo, E.M., Radicati Di Bbrozolo, F., tional tholeiitic basalts in their mineralogy (norma- Rita, F., 1981. Oligocene transitional tholeiitic magmatism in northern tive) and geochemistry, Turkana (Kenya): comparison with the coeval Ethiopian volcanism. Bulletin of Volcanology, 44–53. (2) these transitional tholeiitic basalts are of Oligocene Bellieni, G., Comin-Chiaramonti, P., Marques, L.S., Melfi, A.J., Nardy, age 30.1 ± 1.2 Ma, therefore belong to the oldest A.J.R., Piccirillo, E.M., Stolfa, D., Roisemberg, A., 1984. High- and

lava sequences of the Cameroon Line, low-TiO2 ﬂood basalts from the Parana´ plateau (Brazil): petrology and G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332 331

geochemical aspects bearing on their mantle origin. Neues Jahrbuch Kampunzu, A.B., Mohr, P., 1991. Magmatic evolution and petrogenesis Mineralogische Abhandlungen 150, 273–306. in the East African Rift System. In: Kampunzu, A.B., Lubala, R.T. Bogaard, P.J.F., Worner, G., 2003. Petrogenesis of basanitic to tholeiitic (Eds.), Magmatism in Extensional Structural Settings. Springer- volcanic rocks from the Miocene Vogelsberg, Central Germany. Verlag, Berlin, pp. 85–136. Journal of Petrology 44, 569–602. Kuepouo, G., 2004. Geology, Petrology and Geochemistry of the Tertiary Burke, K., 2001. Origin of the Cameroon Line of volcano-capped swells. Bana Volcano–Plutonic Complex, Cameroon Line, Central Africa. Journal of Geology 109, 349–362. Ph.D. Thesis, Kobe University, Japan, 301p. Caen-Vachette, M., Tempier, P., Nana, J.M., 1991. Le granite de Lembo Lasserre, M., 1978. Mise au point sur les granitoı¨des dits ‘‘ultimes’’ du (partie du complexe volcano-plutonique de Bana), te´moin du magma- Cameroun. Gisement, pe´trographie et geóchronologie. Bulletin du tisme Tertiaire du Cameroun. Geóchronologie. Bulletin de la Socie´te Bureau de Recherches Geólogiques et Minie`re 2, 145–159. Geólogique de France 3, 497–501. Le Bas, M.J., Le Maitre, R.W., Streckeisen, A., Zanettin, B., 1986. A Cantagrel, J.M., Jamond, C., Lasserre, M., 1978. Le magmatism alcalin de chemical classification of volcanic rocks based on the total alkali-silica la ligne du Cameroun au Tertiaire: donneés geóchronologiques K/Ar. diagram. Journal of Petrology 27, 745–750. Compte Rendu Sommaire Seánces Socie´te´ Geólogique France 6, Lee, D.C., Halliday, A.N., Fitton, J.C., Poli, G., 1994. Isotopic variations 300–303. with distance and time in the volcanic islands of the Cameroon line: Condie, K.C., 2003. Incompatible element ratios in oceanic basalts and evidence for the plume origine. Earth and Planetary Sciences Letters komatiites: tracking deep mantle sources and continental growth rates 123, 119–138. with time. Geochimical, Geophysis, Geosysteme (1). Lundstrom, C.C., Shaw, H.F., Ryerson, F.J., Williams, Q., Gill, J., 1998. Condie, K.C., 2005. High field strength element ratios in Archean basalts: Crystal control of clinopyroxene-melt partitioning in the Di–Ab–An a window to evolving sources of mantle plume. Lithos 79, 491–504. system: implications for elemental fractionations in the depleted Coombs, D.S., 1963. Trends and Affinities of Basaltic Magmas and mantle. Geochimica et Cosmochimica Acta 62, 2849–2862. Pyroxenes as Illustrated on the Diopside-Olivine-Silica Diagram, Marzoli, A., Renne, P.R., Piccirillo, E.M., Francesca, C., Bellieni, G., Special Paper, vol. 1. Mineralogical Society of America, pp. 227–250. Melfi, A.J., Nyobe, J.B., N’ni, J., 2000. The Cameroon Volcanic Line Courtillot, V., Davaille, A., Besse, J., Stock, J., 2003. Three distinct types Revisited: Petrogenesis of continental basaltic magmas from litho- of hotspots in the Earth’s mantle. Earth and Planetary Science Letters spheric and asthenospheric mantle sources. Journal of Petrology 41, 205, 295–308. 87–109. De´ruelle, B., Moreau, C., Nkoumbou, C., Kambou, R., Lissom, J., McKenzie, D., O’Nions, R.K., 1991. Partial melt distribution from Njonfang, E., Nono, A., 1991. The Cameroon Line: A Review. In: inversion of rare earth element concentrations. Journal of Petrology Kampunzu, A.B., Lubala, R.T. (Eds.), Magmatism in Extensional 32, 1021–1091. Tectonic Structural Settings. Springer, Berlin, pp. 274–327. McKenzie, D., O’Nions, R.K., 1995. The source regions of ocean island Fairhead, J.D., Okereke, C.S., 1987. A regional study of the West African basalts. Journal of Petrology 36, 133–160. Rift system in Nigeria and Cameroon and its tectonic interpretation. Menzies, M.A., Kyle, P.R., 1990. Continental volcanism: a crust-mantle Tetctonophysics 143, 141–159. probe. In: Menzies, M.A. (Ed.), Continental mantle, Oxford Mono- Fitton, J.G., 1987. The Cameroon Line, Africa: a comparison between graphs on Geology and Geophysics. Oxford Science Publications, oceanic and continental alkaline volcanism. In: Fitton, J.G., Upton, Clarendon press, pp. 157–177. B.G.J. (Eds.), Alkaline Igneous Rocks, Special Publication, vol. 30. Middlemost, E.A.K., 1989. Iron oxidation ratios, norms and the classi- Geological Society, London, pp. 273–291. fication of volcanic rocks. Chemical Geology 77, 19–26. Fitton, J.G., Dunlop, H.M., 1985. The Cameroon line. West Africa, and Moreau, C., Regnoult, J.M., De´ruelle, B., Robineau, B., 1987. A new its bearing on the origin of oceanic and continental alkali basalt. Earth tectonic model for the Cameroon Line, Central Africa. Tectonophysics and Planetary Science Letters 72, 23–38. 139, 317–334. Fitton, J.G., Saunders, A.D., Norry, M.J., Hardarson, B.S., Taylor, R.N., Morimoto, N., Fabries, J., Ferguson, A.K., Ginzburg, I.V., Boss, M., 1997. Thermal and chemical structure of the Iceland plume. Earth and Seifert, F.A., Zussman, J., Aoki, K., Gottardi, G., 1988. Nomenclature Planetary Science Letters 153, 197–208. of pyroxenes. American Mineralogist 73, 1123–1133. Fosso, J., Meńard, J.J., Bardinzeff, J.M., Wandji, P., Tchoua, F.M., Moundi, A., 2004. Les basalts des plateaux du plateau Bamoun: Bellon, H., 2005. Les laves du mont Bangou: une premie`re manifes- Pe´trologie-geóchimie et geóchronology. Implications sur les sources tation volcanique Eoce`ne a affinite´ transitionnelle de la ligne de des magmas, contexte et e´volution geódynamiques. The`se de Doctorat Cameroun. Comptes Rendus des Geósciences 337, 315–325. d’Etat, Universite´ de Yaounde´ I, Cameroun, 208p. Frey, F.A., Garcia, M.O., Wise, W.S., Kennedy, A., Gurriet, P., Albarede, Moundi, A., Meńard, J.J., Reusser, E., Tchoua, F.M., Dietrich, J., 1996. F., 1991. The evolution of Mauna Kea volcano, Hawaii: petrogenesis Dećouverte de basalts transitionnels dans le secteur continental de la of tholeiitic and alkalic basalts. Journal of Geophysical Research 96, Ligne du Cameroun (Massif du Mbam, Ouest-Cameroun). Comptes 14347–14375. Rendus de l’Academie des Sciences, Paris 32 (IIa), 831–837. Green, D.H., 1971. Compositions of basaltic magmas as indicators of Nana, J.M., 1988. Le complexe volcano-plutonique de Bana (Ouest conditions of origin: application to oceanic volcanism. Philanthropic Cameroun). Geólogie et pe´trologie. The`se de Doctorat. Universite´ de Transactions of the Royal Society of London, series A 268, 707–725. Paris XI, France. Halliday, A.N., Dickin, A.P., Fallick, A.E., Fitton, J.G., 1988. Mantle Parsons, I., Brown, W.L., Jacquemin, H., 1986. Mineral chemistry and dynamics: a Nd, Sr, Pb and O isotopic study of the Cameroon Line crystallization conditions of the Mboutou layered gabbro-syenite- volcanic Chain. Journal of Petrology 29, 181–211. granite complex, North Cameroon. Journal of Petrology 27, 1305– Halliday, A.N., Davidson, J.P., Holden, P., DeWolf, C.P., Lee, D.C., 1329. Fitton, J.G., 1990. Trace element fractionation in plumes and origin of Pearce, J.A., Cann, J.R., 1973. Tectonic setting of basic volcanic rocks HIMU mantle beneath the Cameroon Line. Nature 347, 523–528. determined using trace element analysis. Earth and Planetary Science Hill, E., Wood, B.J., Blundy, J.D., 2000. The effect of Ca-Tschermaks Letters 19, 290–300. component on trace element partitioning between clinopyroxene and Plomerova, J., Babuska, V., Dorbath, L., Lillie, R., 1993. Deep silicate melt. Lithos 53, 203–215. lithospheric structure across the Central African shear zone in Irvine, T.N., Baragar, W.R.A., 1971. A guide to the chemical classification Cameroon. Geophysical Journal International 115, 381–390. of the common rocks. Canadian Journal of Earth Sciences 8, 523– Poudjom Djomani, Y.H., Nnange, J.M., Diament, M., Ebinger, C.J., 548. Fairhead, J.D., 1995. Effective elastic thickness and crustal thickness Kampunzu, A.B., Lubala, R.T. (Eds.), 1991. Magmatism in extensional variations in west-central Africa inferred from gravity data. Journal of structural settings. Springer-Verlag, Berlin, pp. 85–136. Geophysical Research 100, 2204–2207. 332 G. Kuepouo et al. / Journal of African Earth Sciences 45 (2006) 318–332

Poudjom Djomani, Y.H., Nnange, J.M., Diament, M., 1997. Lithospheric Thompson, R.N., 1982. Magmatism of the British Tertiary of British structure across the Adamawa plateau (Cameroon) from gravity Tertiary volcanic province. Scottish Journal of Geology 18, 49– studies. Tectonophysics 273, 317–327. 107. Rankenburg, K., Lassiter, J.C., Brey, G., 2005. The role of continental Tilley, C.E., 1950. Some aspects of magmatic evolution, Quart. Journal of crust and lithospheric mantle in the genesis of Cameroon volcanic line the Geophysical Society London 106, 37–61. lavas: constraints from isotopic variations in lavas and megacrysts Weaver, B.L., 1991. The origin of ocean island basalt end-member from the Biu and Jos Plateau. Journal of Petrology 46, 169–190. compositions: trace element and isotopic constraints. Earth and Sato, H., Aramaki, S., Kusakabe, M., Hirabayashi, J.I., Sano, Y., Nojiri, Planetary Science Letters 104, 381–397. Y., Tchoua, F., 1990. Geochemical difference of basalts between Weaver, S.D., Sceal, J.S.C., Gibson, I.L., 1972. Trace element data polygenetic and monogenetic volcanoes in the central part of the relevant to the origin of trachytic and pantelleritic lavas in the East Cameroon volcanic Line. Geochemical Journal 24, 357–370. African Rift System. Contributions to Mineralogy and Petrology 36, Sun, S.S., McDonough, N.F., 1989. Chemical and isotopic systematics of 181–194. oceanic basalts: implications for mantle composition and processes. In: Winchester, J.A., Floyd, P.A., 1976. Geochemical magma type discrim- Saunders, A.A., Norry, M.J. (Eds.), Magmatism in Ocean Basins, vol. ination: application to altered and metamorphosed basic igneous 42. Geological Society Special Publication, pp. 313–345. rocks. Earth and Planetary Science Letters 28, 459–469. Tatsumi, Y., Eggins, S., 1995. Subduction Zone Magmatism. Blackwell, Zanettin, B., Gregnanin, A., Justin-Visentin, E., Piccirillo, E.M., 1980. Oxford. Correlations among Ethiopian volcanic formations with special Tatsumi, Y., Hamilton, D.L., Nesbitt, R.W., 1986. Chemical character- references to the chronological and stratigraphical problems of the istics of fluid phase released from a subducted lithosphere and origin of hTrap Seriesi. In: Geogynamic Evolution of the Afro-Arabian Rift arc magmas: evidence from high-pressure experiments and natural System, Accademia Naz. Dei Lincei, Atti dei Convegni Lincei, vol. 47, rocks. Journal of Volcanology and Geothermal Research 29, 293–309. pp. 23–252. Thirlwall, F.M., Upton, B.G.J., Jenkins, C., 1994. Interaction between Zindler, A., Staudigel, H., Batiza, R., 1984. Isotope and trace element continental lithosphere and Iceland plume-Sr–Nd–Pb isotope geo- geochemistry of young Pacific seamounts: implications for the scale of chemistry of Tertiary basalts, NE Greenland. Journal of Petrology 35, upper mantle heterogeneity. Earth and Planetary Science Letters 70, 839–879. 175–195.